Brooklyn Law School - SSRN papers

University of Pennsylvania Law School Public Law and Legal Theory Research Paper Series Research Paper No. #09-27 ~and~

Brooklyn Law School Legal Studies Research Paper No. 171

Competing Theories of Blackmail: An Empirical Research Critique Of Criminal Law Theory Paul H. Robinson University of Pennsylvania Michael T. Cahill Brooklyn Law School Daniel M. Bartels Columbia University Graduate School of Business

89 Tex. L. Rev. 291 (2010) This paper can be downloaded without charge from the Social Science Research Network Electronic Paper collection: http://ssrn.com/abstract=1477400

Electronic copy available at: http://ssrn.com/abstract=1477400

Brooklyn Law School Legal Studies Research Papers Accepted Paper Series Research Paper No. 171

April 2011

Competing Theories of Blackmail: An Empirical Research Critique Of Criminal Law Theory Paul H. Robinson Michael T. Cahill Daniel M. Bartels

This paper can be downloaded without charge from the Social Science Research Network Electronic Paper Collection: http://ssrn.com/abstract=1477400


Competing Theories of Blackmail: An Empirical Research Critique of Criminal Law Theory Paul H. Robinson,* Michael T. Cahill** & Daniel M. Bartels*** The crime of blackmail has risen to national media attention because of the David Letterman case, but this wonderfully curious offense has long been the favorite of clever criminal law theorists. It criminalizes the threat to do something that would not be criminal if one did it. There exists a rich literature on the issue, with many prominent legal scholars offering their accounts. Each theorist has his own explanation as to why the blackmail offense exists. Most theories seek to justify the position that blackmail is a moral wrong and claim to offer an account that reflects widely shared moral intuitions. But the theories make widely varying assertions about what those shared intuitions are, while also lacking any evidence to support the assertions. This Article summarizes the results of an empirical study designed to test the competing theories of blackmail to see which best accords with prevailing sentiment. Using a variety of scenarios designed to isolate and test the various criteria different theorists have put forth as “the” key to blackmail, this study reveals which (if any) of the various theories of blackmail proposed to date truly reflects laypeople’s moral judgment. Blackmail is not only a common subject of scholarly theorizing but also a common object of criminal prohibition. Every American jurisdiction criminalizes blackmail, although there is considerable variation in its formulation. The Article reviews the American statutes and describes the three general approaches these provisions reflect. The empirical study of lay intuitions also allows an assessment of which of these statutory approaches

* Colin S. Diver Professor of Law, University of Pennsylvania Law School. We thank Mitch Berman, Leo Katz, George Fletcher, and Jim Lindgren for helping us better understand their theories for this study. We also thank Adam Kolber for his useful contributions and Matthew Majarian, University of Pennsylvania Law School Class of 2011, for exceptional research assistance. ** Associate Dean for Academic Affairs and Professor of Law, Brooklyn Law School. Work on this Article was supported by a summer research stipend from Brooklyn Law School, for which Professor Cahill thanks the School and Dean (now President) Joan G. Wexler. Professor Cahill also thanks Paul Means, Rosalyn Scaff, Christopher Serkin, Nelson Tebbe, Ben Trachtenberg, Mary Willis White, and Saul Zipkin for “field testing” our survey instrument and its manipulation checks to help ensure the scenarios were presented in a way that captured the relevant distinctions. ***Assistant Professor of Marketing, Columbia University Graduate School of Business.


292

Texas Law Review

[Vol. 89:291

(if any) captures the community’s views, thereby illuminating the extent to which existing law generates results that resonate with, or deviate from, popular moral sentiment. The analyses provide an opportunity to critique the existing theories of blackmail and to suggest a refined theory that best expresses lay intuitions. The present project also reveals the substantial conflict between community views and much existing legislation, indicating recommendations for legislative reform. Finally, the Article suggests lessons that such studies and their analyses offer for criminal law and theory. I.

Competing Theories of Blackmail ...................................................... 295 A. Theories of Blackmail as a Crime Against the Threat Recipient .... 296 1. The Wrongful Intention Theory................................................... 296 2. The Continuing Domination Theory ........................................... 299 B. Theories of Blackmail as a Crime Against Third Parties ................ 300 1. The Breach-of-Duty Theory ........................................................ 300 2. The Usurping Authority Theory .................................................. 303 C. Theories of Blackmail as a Societal Harm ...................................... 304 D. Theories of Blackmail as Noncrime: The Abolitionist Position ..... 306 II. Statutory Approaches .......................................................................... 308 III. Testing Community Views ................................................................. 313 A. Design and Methodology ................................................................ 314 B. Theory Predictions .......................................................................... 322 1. Wrongful Intention ...................................................................... 323 2. Breach of Duty ............................................................................ 325 3. Continuing Domination .............................................................. 327 4. Usurping Authority ..................................................................... 328 5. Efficient Information Allocation ................................................. 330 C. Statutory Liability Patterns ............................................................. 331 D. Results and Discussion ................................................................... 335 1. Manipulation-Check Results ....................................................... 335 2. Liability Results .......................................................................... 336 3. The Theories ............................................................................... 339 4. The Statutory Schemes ................................................................ 342 5. More Sophisticated Measures of Best Fit with Subjects’ Criminalization Views................................................................. 343 6. Effect of Seriousness of Threat and Wrongfulness of Victim’s Undisclosed Conduct .................................................................. 346 IV. Conclusion .......................................................................................... 347 Appendix A: Scenario Legal Liability Analyses ........................................ 350


2010]

Competing Theories of Blackmail

293

The crime of blackmail has risen to national attention as a result of the highly public scandal involving David Letterman.1 Yet as titillating or colorful as the details of one notorious case may be, they hardly provide the only reason to take an interest in this wonderfully curious crime. Indeed, blackmail has long been the favorite offense of clever criminal law theorists. It criminalizes the threat to do something that would not be criminal if one did it. If your acquaintance is having an affair, it is no crime to tell his wife of his infidelity. However, if you threaten to do so unless he pays you $100, that threat is criminal—even if he would consider it a bargain and quickly accept your offer. Unlike the similar but uncontroversial category of extortion—which involves conditional threats to engage in criminal acts, such as a threat to injure someone unless paid—this disparate legal treatment of the threat and the threatened activity makes blackmail seem like a puzzle or, in a well-known and often-repeated characterization, a “paradox.”2 Though blackmail is not extortion, something about the use of coercion might seem to comprise the gravamen of the offense. But then, many forms of coercion are not criminal. A source pressed by a reporter to provide information or else face an unflattering portrayal, or an employer pressured to either give her best salesman a raise or watch him quit, may feel as much coercion as the recipient of a blackmail threat, yet this coercion is not criminal. What is it about the nature or circumstances of a threat that make it blackmail and not mere “hard bargaining”? There exists a rich literature on the issue, with many prominent legal scholars jumping in to offer their accounts.3 Each theorist has his own expla1. See Bill Carter & Brian Stelter, Extortion Case Raises Questions for Letterman and His Network, N.Y. TIMES, Oct. 3, 2009, at A1 (describing Robert Halderman’s attempt to obtain two million dollars from Letterman in exchange for not revealing Letterman’s lurid sexual history); Bill Carter, Inside CBS, Disbelief at an Arrest, N.Y. TIMES, Oct. 8, 2009, at B1 (chronicling CBS employees’ astonished reactions to the Halderman allegations); Lizzie Widdicombe, Brainteaser: You’ve Got Mail, NEW YORKER, Oct. 19, 2009, at 28 (using the Letterman case as springboard to a discussion of the intellectual riddle of blackmail). 2. See, e.g., James Lindgren, Unraveling the Paradox of Blackmail, 84 COLUM. L. REV. 670, 671 (1984) (contemplating the paradox that a blackmailer combines a legal means and a legal end to achieve an illegal result). 3. See, e.g., 4 JOEL FEINBERG, THE MORAL LIMITS OF THE CRIMINAL LAW: HARMLESS WRONGDOING 240 (1988) (noting that “the exploitation principle provides a rationale for blackmail laws where the liberal’s unsupplemented harm principle finds only a ‘paradox’”); Mitchell N. Berman, The Evidentiary Theory of Blackmail: Taking Motives Seriously, 65 U. CHI. L. REV. 795, 797–98 (1998) (proffering a theory of blackmail where the overt act of blackmail serves only the evidentiary function of helping a fact finder separate disclosures of embarrassing information based on acceptable motivations from disclosures based on unacceptable motivations); Ronald H. Coase, The 1987 McCorkle Lecture: Blackmail, 74 VA. L. REV. 655, 674–75 (1988) (noting a variety of distinctions between blackmail and business negotiations, such as the fact that instances of blackmail are not constrained by market competition or the hope of future business relations); Richard A. Epstein, Blackmail, Inc., 50 U. CHI. L. REV. 553, 566 (1983) (arguing that blackmail is

294

Texas Law Review

[Vol. 89:291

nation as to why the blackmail offense exists. Some theories are instrumentalist, explaining the criminalization of blackmail solely in terms of its positive practical effects rather than the wrongfulness of the underlying conduct.4 Many theories, however, seek to justify the position that blackmail is a moral wrong.5 It is typical for such theories to defend their moral judgments or assertions by relying on the claim that a stated moral position accords with widely shared moral intuitions. Indeed, the standard methodology for these blackmail theories is to seek a “reflective equilibrium” between general normative principles and shared intuitions about the proper outcome of particular cases.6 Blackmail theories thus place considerable reliance on claims about lay intuitions. Yet different theorists make different claims about “our” shared moral judgments regarding particular blackmail scenarios and do so without offering empirical data to support their favored intuition or to refute any other proffered intuition. So who is right? Which theory, if any, accurately captures people’s shared moral intuitions about the contours of blackmail? Or are there no such shared intuitions at all? This Article summarizes the

criminalized because the demand is usually “part of an overall scheme of abuse, itself rife with coercive and fraudulent elements,” and that “[b]lackmail is made a crime not only because of what it is, but because of what it necessarily leads to”); George P. Fletcher, Blackmail: The Paradigmatic Crime, 141 U. PA. L. REV. 1617, 1626 (1993) (suggesting that “the proper test [for whether an act should constitute blackmail] . . . is whether the transaction with the suspected blackmailer generates a relationship of dominance and subordination”); Douglas H. Ginsburg & Paul Shechtman, Blackmail: An Economic Analysis of the Law, 141 U. PA. L. REV. 1849, 1873 (1993) (arguing that blackmail prohibition is an “economically rational rule” because “[i]f such threats were lawful, there would be an incentive for people to expend resources to develop embarrassing information about others in the hope of then selling their silence”); Leo Katz, Blackmail and Other Forms of Arm-Twisting, 141 U. PA. L. REV. 1567, 1615 (1993) (defining blackmail as a situation in which the circumstances lead the victim to “prefer to be subjected to a greater rather than a lesser wrong”); Jeffrie G. Murphy, Blackmail: A Preliminary Inquiry, 63 MONIST 156, 163–66 (1980) (turning to the social policies of preventing invasions of privacy, protecting the free press from being outpriced by private negotiation, and increasing the availability of information on public officials to justify the distinction between legal economic transactions and illegal instances of blackmail); Richard A. Posner, Blackmail, Privacy, and Freedom of Contract, 141 U. PA. L. REV. 1817, 1820 (1993) (stating that blackmail “diminishes social wealth” and is a “sterile redistributive activity” comparable to “(simple) theft”); Steven Shavell, An Economic Analysis of Threats and Their Illegality: Blackmail, Extortion, and Robbery, 141 U. PA. L. REV. 1877, 1903 (1993) (stating that “there is still an obvious incentive-based reason for making blackmail illegal: to avoid being blackmailed by [persons] who might by chance be present, potential victims will exercise excessive precautions or reduce their level of innocent, yet embarrassing, activities”). 4. See, e.g., Ginsburg & Shechtman, supra note 3, at 1850 (applying an economic analysis to the criminalization of blackmail and finding the criminalization consistent with economic rationality). 5. See, e.g., Berman, supra note 3, at 798 (“[S]ociety can punish the blackmailer . . . because the [blackmailer] causes (or threatens) harm while acting with morally culpable motives.”). 6. Mitchell N. Berman, Blackmail (manuscript at 6, 7 & n.7) (on file with authors) (embracing the reflective equilibrium approach and expressing the belief “that most blackmail theorists share these methodological commitments”), in OXFORD HANDBOOK ON THE PHILOSOPHY OF THE CRIMINAL LAW (John Deigh & David Dolinko eds., forthcoming 2010).

2010]


295

results of an empirical study designed to test the various competing moral theories of blackmail to see which best accords with prevailing sentiment. Part I reviews the alternative theories, while Part III compares these to the results of an empirical test of lay intuitions. Blackmail is not only a common subject of theoretical discussion but a common object of criminal prohibition. Every American jurisdiction criminalizes blackmail, although there is considerable variation in statutory formulation. Part II reviews the American statutes and describes the three general approaches these provisions reflect. The empirical study of lay intuitions reported in Part III also allows an assessment of which of these statutory approaches, if any, captures the community’s views. The analyses provide an opportunity to evaluate and critique the existing theories of blackmail and ultimately, perhaps, to develop a refined theory that best expresses lay intuitions. The present project also reveals the substantial conflict between community views and much existing legislation and indicates possible avenues for legislative reform. The Article’s conclusion suggests lessons that this study offers and that other similar studies might offer for criminal law and theory. I.


Theories of the proper basis for the criminal prohibition against blackmail differ profoundly from one another. One reason for this is that, unlike many other crimes, it is not entirely clear whom (if anyone) blackmail harms or victimizes. On one level, the “victim” of a blackmail threat seems to be the person receiving the threat who is forced to pay money (or give up something else of value) to prevent the blackmailer from carrying out the threat. Some theories of blackmail, which we discuss in subpart A, are based on this notion that the threat recipient is properly seen as the crime’s victim. Yet being blackmailed is arguably less harmful to that “victim” than if the blackmailer were simply to perform the threatened activity without first making the threat, as the blackmailer is entirely at liberty to do. If the recipient of the threat accedes to the blackmailer’s demand, presumably he does so because he finds that preferable to having the blackmailer carry out the threat, as would occur if the demand were rejected. He is therefore in a better situation by virtue of having the option to pay than he would be if the blackmail threat—or, as it could also be seen, the blackmail offer—were never made. Some theories, following this logic, conclude that the true victim of blackmail is some other party who, because of the blackmail transaction, is losing access to what the blackmailer would provide that party (typically information) if the blackmailer could not extract value from the threat recipient by engaging in blackmail instead. Subpart B discusses these theories. Some theories do not depend on any claim that individual cases of blackmail necessarily harm anyone at all. Rather, they defend the criminali-

296

Texas Law Review

[Vol. 89:291

zation of blackmail on the ground that if blackmail were legal, some overall social harm would ensue, such as a general loss of privacy or an inefficient allocation of resources to investments in protecting secrets.7 We did not include such theories in our empirical survey because their premises are such as to make them unconcerned with whether any particular case of blackmail, or even the practice of blackmail as a whole, is wrongful or merits blame. Accordingly, these theories make no claim about relying on (or being able to predict) popular views as to when blackmail deserves punishment. Even without their explicit reliance on lay intuitions of justice to justify themselves, we might nonetheless be interested to see whether such economic theories accord or conflict with lay intuitions. However, the nature of most such theories, at least as expressed in the current literature, lacks sufficient content to actually formulate an offense. That is, these theories may offer a basic explanation of why some form of blackmail offense should exist, but they typically do not tell us with any precision what such an offense should look like.8 We discuss these theories more fully in subpart C. Finally, there is the position that the criminalization of blackmail lacks any sound basis and is therefore inappropriate. We discuss this abolitionist position in subpart D. A. Theories of Blackmail as a Crime Against the Threat Recipient Two major theories of blackmail see it as fundamentally a crime that victimizes the person being threatened. The first view, set forth at different times and in somewhat different variations by Mitchell Berman and Leo Katz, claims that blackmail is truly a species of extortion, i.e., a threat to engage in a wrongful act. Accordingly, the putative blackmail “paradox” vanishes because both the act and the threat are wrongful. The second view, espoused by George Fletcher, finds blackmail to be a harm to the recipient not by virtue of the threat per se but because of the threat’s potential for repetition, which creates the possibility that the recipient will be forced into an ongoing relationship of subordination to the blackmailer. 1. The Wrongful Intention Theory.—Over a series of articles, Mitchell Berman has elaborated and slightly refined what he calls the “evidentiary”

7. See, e.g., Epstein, supra note 3, at 566 (“[T]he [blackmail] demand will not take place in isolation, but will be part of an overall scheme of abuse, itself rife with coercive and fraudulent elements. . . . Blackmail should be a criminal offense even under the narrow theory of criminal activities because it is the handmaiden to corruption and deceit.”); Posner, supra note 3, at 1832 (“In the face of this uncertainty [if blackmail were legal], the safest guess is that allowing the blackmailing . . . would yield a net social loss equal to the resources expended in blackmailing and in defending against blackmailing [to protect secrets].”). 8. Given this limitation, one may wonder about the value of such theories as they relate to the development of substantive criminal law.

2010]


297

theory of blackmail.9 For Berman, blackmail is wrongful if and only if it would be wrong for the blackmailer to carry out the threatened act.10 Where the threatened act is inherently harmful or wrongful, as where one threatens to injure another unless paid, the threat presents a routine case of extortion, for which it is relatively easy to justify assigning blame or punishment. Yet even where the threatened act is not wrongful per se, it might also be wrongful based on the actor’s culpability in performing it—and this, Berman claims, is the case with blackmail.11 Further, Berman contends that the threat itself provides evidence of the blackmailer’s culpability, i.e., evidence that he knows carrying out the threat will harm the recipient and that, were he to carry out the threat, he would be doing so for that very reason. In earlier writings, Berman described the offender’s culpability as rooted in his motivations;12 in more recent work, Berman discusses culpability in terms of the offender’s knowledge and beliefs.13 Berman’s explicit goal for his theory is that it should track common intuitions regarding blackmail as closely as possible: he is engaged in a process of “reflective equilibrium” in which the general theory is meant to track general intuitive reactions to specific cases.14 (Berman also thinks this process of rationalizing and seeking to track common moral intuitions is the norm for blackmail theorists.15) Somewhat like Berman, Leo Katz advances a test for blackmail that asks whether the threatened activity is itself wrongful.16 The harm of blackmail for Katz is in forcing the recipient of the threat to choose between two “immoralities,” namely, facing the prospect of (1) having to pay for the blackmailer’s silence or (2) having the blackmailer carry out the threat.17 Importantly, and again similar to Berman’s view, carrying out the threat

9. See Berman, supra note 3, at 848–51 (summarizing the evidentiary theory); Mitchell N. Berman, Meta-Blackmail and the Evidentiary Theory: Still Taking Motives Seriously, 94 GEO. L.J. 787, 789 (2006) [hereinafter Berman, Meta-Blackmail] (arguing that the evidentiary theory explains why blackmail is coercive); Mitchell N. Berman, Book Review, On the Moral Structure of White Collar Crime, 5 OHIO ST. J. CRIM. L. 301, 322–25 (2007) [hereinafter Berman, White Collar Crime] (putting forth an abridged version of the evidentiary theory); Berman, supra note 6 (manuscript at 58–59) (discussing how the evidentiary theory relates to morality). 10. Berman, supra note 6 (manuscript at 36). 11. Berman, White Collar Crime, supra note 9, at 323. 12. Berman, supra note 3, at 839–40; Berman, Meta-Blackmail, supra note 9, at 791. 13. See Berman, supra note 6 (manuscript at 55, 56 n.118) (dispelling Michael Gorr’s blackmail approach by discussing the importance of the actor’s belief and knowledge regarding what he “morally ought to do” in the blackmail puzzle). 14. See id. (manuscript at 6–7) (discussing individual intuitions in terms of blackmail). 15. Id. (manuscript at 7 & n.7). 16. See Katz, supra note 3, at 1599 (stating that blackmail requires a threat of at least mildly wrongful conduct). 17. Id. at 1598.

298

Texas Law Review

[Vol. 89:291

might be “immoral” not only because of its objective harmfulness but also because of the blackmailer’s motivations—such as a spiteful or vindictive decision to expose a secret in retaliation for not having one’s demand satisfied.18 Like Berman, Katz explicitly relies on moral intuitions he expects readers to share and seeks to generate a theory of blackmail that accords with those intuitions. For example, he rejects Richard Epstein’s social-harm account of blackmail on the ground that it does not reflect “our instinctive revulsion at the practice.”19 He summarizes his own account as capturing conduct that is “deemed by us a very major wrong.”20 Both Berman and Katz, then, think blackmail is wrongful if and only if carrying out the threat would be wrongful. They also have detailed and nuanced views of what would make a threat, or any other conduct, wrongful— and on this broader score, their views sometimes (though not often) differ from one another. Importantly, however, their broader views about wrongfulness are not directly relevant to the project at hand. The present task is to determine the extent to which their view of when and why blackmail specifically is wrongful tracks common lay views of that same issue. As to blackmail in particular, Berman and Katz take the same position: the wrongfulness of blackmail depends on the wrongfulness of the threatened act. Significantly, this view of blackmail could be “right” (in the sense of tracking lay intuitions) even if neither Berman nor Katz is right in his broader positions as to what makes actions wrongful. In other words, if laypeople consistently give the same answer to the questions (1) “is this blackmail?” and (2) “would carrying out this threat be wrongful?” then they are employing the Berman/Katz approach to blackmail, even if they do not share Berman’s or Katz’s views regarding why carrying out the threat would be wrongful. If the driving criterion behind lay assessments of blackmail is 18. See id. (noting the case where a threatened act “is immoral only because, if it were to be done, it would be done for purely retaliatory reasons—retaliation for [the victim’s] refusal to pay”); id. at 1600 (discussing the nonhiring of a job applicant who refuses to have sex with her employer as wrongful because it would be retaliatory); id. at 1602. Katz has a similar response when addressing the situation of reporting information to the IRS out of a retaliatory motivation: Feinberg is incorrect about such cases as the proposal to withhold damaging information from the IRS, because a retaliatory reporting of such information to the IRS, (i.e., the reporting of such information not to help the government, but to settle a score) strikes us as quite immoral, not immoral at the level of criminality or tortiousness, but immoral all the same. Leveraging such immoral conduct into a substantial gain then becomes blameworthy at the level of theft. Id. 19. Id. at 1578; see also id. at 1580 (finding fault with Feinberg’s theory because it fails to include a case that “is viewed by many as the quintessential blackmail case”); id. at 1581 (assessing Lindgren’s theory by noting that it “pretty closely matches our intuitions at the descriptive level, although it seems perhaps a bit underinclusive” as it fails to “account for several cases which many would agree clearly reek of blackmail”). 20. Id. at 1615.

2010]


299

rooted in a moral assessment of the wrongfulness of carrying out the threat, then lay intuitions agree with Berman and Katz about blackmail, even if they disagree about other aspects of moral theory. 2. The Continuing Domination Theory.—George Fletcher has put forth a theory according to which blackmail is wrongful because it creates a relationship of “dominance and subordination” between the blackmailer and the recipient of the threat.21 What distinguishes blackmail from other situations of hard bargaining between parties to a transaction is the potentially ongoing duration of the threat: it may involve not just one demand of money for silence but repeated demands because the blackmailer remains privy to the damaging information and can continue to extract money or other value from the threatened party. Like Berman, Fletcher aims to track shared views of what constitutes blackmail, explicitly seeking “reflective equilibrium” between theory and intuition, which Fletcher describes as “requir[ing] a convincing fit between . . . agreed-upon outcomes . . . and general principles that can account for these outcomes.”22 However, Fletcher does not fully address whether the threatened act must involve a certain degree of coercion or whether the blackmailer’s demand must reach a certain magnitude for the threat to create a relation of dominance and subordination. As to the first of these, Fletcher seems to answer in the negative because he thinks a proposal can be viewed as blackmail whether it is considered a “threat” or an “offer.”23 Elsewhere Fletcher seems to suggest that only certain kinds of demands qualify as blackmail, however, because he asserts that “no one can dominate someone else by asking for money to do or not to do that which is in one’s recognized domain of freedom. . . . [T]here is no blackmail in demanding payments to do or not to do that which one has a right to do.”24 Yet that statement surely cannot be accurate as written, for any classic case of informational blackmail presents a situation where the blackmailer has the freedom or “right” to disclose the information rather than seeking payment, and the recipient of the threat has no legal or moral “right” to prevent that disclosure (in the case of disclosure of a crime, quite the contrary). It is equally clear that Fletcher himself views such cases as blackmail.25 As to the magnitude of the demand, Fletcher 21. See Fletcher, supra note 3, at 1626–29 (proposing a dominance-and-subordination test and applying it to a set of paradigmatic situations). 22. Id. at 1617. 23. See id. at 1623 (“I am skeptical about whether a coherent account is available for these parallel distinctions between threats and offers and between nonproductive and productive exchanges.”). 24. Id. at 1627–28. 25. See, e.g., id. at 1617–19 (describing ten paradigmatic hypothetical situations, including, inter alia, criminal and noncriminal informational blackmail situations).

300

Texas Law Review

[Vol. 89:291

asserts only that a “minimal” demand does not reach the level of blackmail, though he acknowledges that “[e]xactly what is required . . . is not clear.”26 B. Theories of Blackmail as a Crime Against Third Parties Two other theories of blackmail view it not as a crime against the recipient of the threat but as a crime against whomever would have received the blackmailer’s information had the blackmail not taken place. Joel Feinberg conceptualizes this view in terms of the moral duties of the blackmailer to the other party. James Lindgren describes it in terms of the third party’s authority to regulate or discipline the threat recipient directly, which the blackmailer is usurping for personal gain. 1. The Breach-of-Duty Theory.—Joel Feinberg’s test for blackmail asks whether, even holding aside the existence of the threat, the blackmailer’s disclosure or nondisclosure of the information would be wrongful.27 If the blackmailer has a duty to disclose the information, then it is improper for the blackmailer to violate that duty and keep silent in exchange for money. For example, withholding disclosure of the threatened party’s criminal activity would be wrongful because one has a moral duty to report crimes; hence, nondisclosure of criminal activity in return for money would be blackmail.28 On the other hand, if the blackmailer has a duty not to disclose the information, then it is improper to threaten disclosure.29 (Such threats fall within the category of extortion—the threat to do an act that is itself impermissible—whose prohibition is relatively noncontroversial.) Yet where the person is allowed, but not obliged, to disclose the information, the conditional threat to do so is not blackmail for Feinberg.30 In short, Feinberg takes seriously the so-called paradox of blackmail—its apparent willingness to punish a threat to perform conduct that would itself be permissible absent

26. Id. at 1627. 27. See FEINBERG, supra note 3, at 211–13, 238–58 (defining five categories of blackmail-like threats and analyzing their wrongfulness in light of the harm caused to the victim, the harm caused to society, and the unjust gain to the blackmailer). 28. See id. at 241–45 (arguing that members of society do not have a right to withhold reports of a crime because nondisclosure causes a public harm). 29. See id. at 249–58 (noting that one has a civil duty not to make accusations known to be false and arguing that one has a moral duty to refrain from making truthful accusations of past wrongful conduct or conduct that is innocent but embarrassing). 30. See id. at 245–49 (arguing that blackmail should only be criminalized in situations where the threatened disclosure or offered failure to disclose would in itself violate a legal or civic duty); id. at 275 (“I don’t see how a coherent criminal code based on liberal principles . . . can prohibit people from offering, in exchange for consideration, not to do what they have an independent legal right (but no legal duty) to do.”).

2010]


301

the threat—and maintains that law should only sanction the threat where it would also sanction the threatened conduct.31 Whether Feinberg’s theory depends on the breach of a legal duty or a moral duty is not always clear. For example, Feinberg posits a duty to report crimes to the police, though he recognizes that “[t]here is admittedly a problem about the precise status”32—legal or moral—of that duty and concludes that “our political system . . . clearly imposes a civic duty . . . to cooperate with law enforcement, even when that duty is not specifically enforced by the criminal or civil law.”33 Such a “civic duty” is enough for Feinberg to find that its violation in return for pay is blackmail. Feinberg is also willing to allow the criminal law’s blackmail prohibition to rest on duties imposed under civil law,34 such as the tort law governing invasions of privacy.35 Further, Feinberg makes a normative argument that civil law should recognize additional duties, such as a defamation claim for some truthful statements,36 whose threatened violation would then also support blackmail liability.37 It seems fair to say, then, that for Feinberg a threat can be considered blackmail if it implicates the violation of a duty that is or could be legitimately imposed by law. Such a violation could arise from the satisfaction of the blackmailer’s demand (preventing disclosure of information he had a duty to disclose), or its nonsatisfaction followed by the carrying out of the threat (leading to disclosure of information he had a duty not to disclose). This category of legitimate (actual or potential) legal duties is similar to but distinct from the notion of a moral duty. Some behavior might be seen as immoral but outside the proper scope of law; as Feinberg says of the case where one knows of another’s adultery, even if we think that disclosing the adultery might be the right thing to do, “[n]o law requiring or forbidding his disclosure would be justified.”38 Feinberg’s general sense of when

31. See id. at 240, 246 (noting that only some types of blackmail are paradoxical but that criminalization of these types cannot be justified on principles of liberalism); id. at 258 (stating that criminalization in a liberal penal code should only be allowed if it “would not stumble over the paradox of blackmail”); id. at 275 (noting of his argument for decriminalizing certain commonly recognized instances of blackmail, “I came to this radical conclusion only because I take the argument of the ‘paradox of blackmail’ very seriously”). 32. Id. at 243. 33. Id. at 244. 34. See id. at 253 (asserting that “[t]he important point is that ‘the law’ . . . imposes a duty,” not whether the legal duty is criminal or civil). 35. See id. at 250–51 (arguing that the duty to not disclose certain damaging information, as imposed by the tort law of privacy invasions, justifies criminalization of blackmail in such cases). 36. See id. at 254–56 (supporting recognition of such a legal claim). 37. See id. at 254 (“It is open to the liberal, however, to argue that there ought to be a civil remedy for such moral wrongs, so that he can argue for criminalization . . . without being thwarted by the paradox of blackmail.”). 38. Id. at 249.

302

Texas Law Review

[Vol. 89:291

criminalization is justified is driven by a liberal demand for some demonstration that the law would prevent harm or offense to others.39 Yet Feinberg’s discussion of adultery-blackmail, criminalization of which Feinberg personally opposes, also indicates his willingness to defer to shared community judgements about the proper scope of the law’s reach: Surely most of those who advocate criminalization of adulteryblackmail would not also advocate legislation making it an independent crime to inform betrayed spouses; nor would they advocate prior legislation making it a legal duty to inform betrayed spouses. They cannot have it both ways. Either the blackmailer should have a duty to inform (or a duty not to, as the case may be) in which case it would be consistent to prohibit him from threatening to violate that duty unless paid off, or he should have no legal duty one way or the other, in which case it would be incoherent to punish him for threatening to do what is within his legal rights.40 If Feinberg’s empirical assertion about public views as to prohibiting disclosure (or nondisclosure) of adultery were shown to be wrong, presumably he would change his position about the propriety of criminalizing adultery-blackmail. Accordingly, for Feinberg, whether the law should prohibit a threat (i.e., treat it as blackmail) depends on whether the law should also prohibit doing the threatened act. If not—if the person making the threat “should have no legal duty one way or the other”41—then the threat is not blackmail. As with the wrongful-intention theory, then, it is possible for subjects to agree with Feinberg about the proper criterion for blackmail but to disagree about how to employ that criterion in particular cases. For the Berman/Katz theory, it is possible to agree with the position that threats should constitute blackmail if and only if they are “wrongful,” but to disagree with the theorists’ own broader views about what is “wrongful.” So too here, it is possible to disagree with Feinberg as to the proper scope of our underlying legal obligations while agreeing that blackmail should apply only to threats that entail violation of those obligations. Thus, like other theorists, Feinberg makes an appeal to shared intuitions in justifying the criminalization of blackmail as a general matter and his own account of blackmail in particular. He posits the existence of wide consensus as to the wrongfulness of the blackmailer’s conduct: “It is a free-floating evil, many people would judge, that he [the blackmailer] should make a big gain as a byproduct of someone else’s crime or indiscretion, that he should profit

39. See id. at 275 (“I have tried to find a liberal alternative to the legal moralist’s account of blackmail . . . .”). 40. Id. at 249. 41. Id.

2010]


303

unproductively from others’ wrongdoing. That his gain is unjust seems clear.”42 Feinberg also maintains that his own account of blackmail would prohibit “precisely those actions that common sense most insistently demands should be criminal.”43 At the same time, what dictates the contours of Feinberg’s view is not an appeal to popular moral intuition but an effort to develop an understanding of blackmail consistent with liberal principles. Indeed, Feinberg’s desire to embrace such principles and avoid what he considers genuinely “paradoxical” cases of blackmail leads him to develop a theory that excludes cases commonly thought to be paradigmatic examples of blackmail, such as the threat to expose another’s adultery.44 2. The Usurping Authority Theory.—For James Lindgren, the “victim” of blackmail is not the recipient of the threat but some third party whose interests the blackmailer is exploiting or suppressing. In the typical case of a conditional threat to disclose information, the wrong consists of the blackmailer’s usurping or “leveraging” the interests of the party entitled to the information: law enforcement authorities, a wronged spouse, etc.45 In a nutshell, for Lindgren, the wrong of blackmail is that the “blackmailer is negotiating for his own gain with someone else’s leverage or bargaining chips.”46 Lindgren offers this account of blackmail in an effort to track common intuitions, describing his project as seeking to “meaningfully distinguish” between “large classes of threats that nearly everyone agrees ought to be illegal and other large classes of threats that nearly everyone agrees ought to be permitted.”47 Whether the blackmailer is “leveraging” another’s position or “using another’s chips” can be determined by examining the nature of the blackmailer’s demand. If the blackmailer is merely replicating the demand the other party would make if that party had the information, then no blackmail exists.48 Even if the demand is not exactly what the other party would 42. Id. at 239. 43. Id. at 276. 44. See id. at 245–49 (concluding that threats to expose adultery should not be criminalized as blackmail because imposing a duty to reveal, or not to reveal, adultery would be inconsistent with liberal principles). 45. See James Lindgren, Blackmail: An Afterword, 141 U. PA. L. REV. 1975, 1981 (1993) (explaining the theory that “someone who threatens to expose criminality or tortious behavior [is] trading on leverage that properly belongs to others”); Lindgren, supra note 2, at 702 (“What makes [the blackmailer’s] conduct blackmail is that [the blackmailer] interposes himself parasitically in an actual or potential dispute in which he lacks a sufficiently direct interest.”). 46. Lindgren, supra note 2, at 702. 47. Id. at 680. 48. See id. at 714 (asserting there is “no blackmail” if “[t]here is a perfect congruence between the advantage sought . . . and the leverage used”).

304

Texas Law Review

[Vol. 89:291

seek, so long as the maker of the threat is “trying in good faith” to benefit the other party rather than himself, Lindgren expresses “doubt that anyone would consider” the threat to be blackmail.49 But if the person “requests something in return for suppressing the actual or potential interests of others,” the request is blackmail.50 C. Theories of Blackmail as a Societal Harm Some theories supporting the criminalization of blackmail are unconcerned with its moral status. For these theories, the prohibition of blackmail is justified not because blackmail is wrongful but because it is harmful (some would say “costly”) to society. These theories do not focus on the harm an individual act of blackmail might cause the recipient of the threat but on the overall social costs that would arise from the improper behavioral incentives legalized blackmail would create. Some writers view these costs in terms of the unproductive efforts of would-be blackmailers, some in terms of the excessive privacy investments required of would-be blackmailees, but the shared underlying perspective is that blackmail’s legal status should depend on a societal cost–benefit analysis rather than a moral inquiry. A host of commentators have provided such a law-and-economics analysis of blackmail, according to which it is properly criminalized because it leads to inefficient allocation of resources.51 These analyses are interested only in contemplating how legalizing or criminalizing blackmail would affect overall societal incentives to ferret out secrets, to overinvest in security, or to engage in fraud.52 They are uninterested in generating a formulation of 49. Id. at 715. 50. Id. at 672. 51. See, e.g., Coase, supra note 3, at 674 (opining that blackmailers inefficiently expend resources gathering and transacting for the nondisclosure of information); Epstein, supra note 3, at 561, 566 (concluding that blackmail should be criminalized because, while an economic analysis of only the actions comprising the blackmail transaction may seem favorable, a broader analysis would take into account the host of inefficient auxiliary behavior encouraged by blackmail); Ginsburg & Shechtman, supra note 3, at 1873 (concluding that blackmail is economically inefficient because it encourages people to expend resources to gain information to protect information); William M. Landes & Richard A. Posner, The Private Enforcement of Law, 4 J. LEGAL STUD. 1, 26 (1975) (discussing the many ways in which private enforcement of laws through blackmail would incentivize inefficient behavior like fabricating evidence or entrapping victims); Richard H. McAdams, Group Norms, Gossip, and Blackmail, 144 U. PA. L. REV. 2237, 2292 (1996) (demonstrating that a ban on blackmail combined with social norms will produce the most efficient distribution of information); Posner, supra note 3, at 1818 (arguing that, while blackmail is a voluntary transaction, it should be prohibited because it is on average wealth reducing); Shavell, supra note 3, at 1902 (explaining that economic analysis supports criminalizing blackmail because of blackmail’s tendency to incentivize wasteful gathering and protecting of information). 52. See, e.g., Epstein, supra note 3, at 564–65 (pointing out that the opportunity for legalized blackmail will give blackmailers an incentive to help their victims perpetuate fraud); Shavell, supra note 3, at 1894–95 (analyzing the effect of blackmail on the incentive to expend effort to obtain information and to take preventative measures to avoid being blackmailed).

2010]


305

blackmail that achieves the “right” or fair result in individual cases. Rather, according to this approach and its underlying concerns, the quality of a given formulation of blackmail would be assessed in terms of its likelihood of achieving the “optimal” or efficient overall societal level of blackmail, i.e., the level at which the marginal costs of preventing blackmail exceeded the marginal costs of the blackmail itself.53 They are not concerned with the morality vel non of blackmail—indeed, some of these accounts raise the suggestion that blackmail might also have some efficiency-related benefits as a form of private law enforcement.54 Like the law-and-economics theorists, Jeffrie Murphy has offered an account focused on the general incentives that would arise if blackmail were legal.55 Murphy’s account is slightly distinct in its concern with the value of individual privacy.56 While the economic account is mostly concerned with the prospect of wasted or unproductive efforts to obtain information, Murphy is concerned with the likelihood of privacy invasions and seems to think that such invasions are inherently problematic.57 At the same time, Murphy would grant an exception allowing the blackmail of public figures, presumably because the information obtained from investigations of public figures would be more socially useful than information about purely private figures.58 In the end, then, Murphy seems to be addressing a set of cost– benefit tradeoffs involving investments in obtaining or protecting secret information, similar to the typical economic theory. Theorists of this type typically set themselves the task of merely justifying the blackmail prohibition’s existence rather than specifying its proper scope. For economists, voluntary transactions, seemingly undertaken for mutual gain, are generally thought to be desirable, and their prohibition, as in the criminalization of blackmail, therefore stands in need of some explanation.59 Having given the explanation, however, law-and-economics 53. See Landes & Posner, supra note 51, at 42–43 (applying the concept of economically efficient private enforcement of laws to comment on the status of blackmail and concluding that society permits the private enforcement of blackmail-like demands where additional public enforcement would, according to broad social norms, not be worth the expenditure associated with the additional enforcement). 54. Id. (suggesting that blackmail by private individuals can substitute for public law enforcement because the amount that the blackmailed person should be willing to pay is equal to the cost of the penalty that law enforcement would impose). 55. See generally Murphy, supra note 3 (discussing possible incentive-based justifications for a prohibition against blackmail). 56. See id. at 159–60, 163–66 (stating that “the protection of privacy does play a role in justifying the criminalization of blackmail” and discussing different privacy issues). 57. See id. at 159 (arguing that a blackmailer acts wrongly “not because he is simply proposing an unjust economic transaction, but because he is economizing a part of life which he has no right to economize”). 58. Id. at 164. 59. See supra note 51 and accompanying text.

306

Texas Law Review

[Vol. 89:291

accounts of blackmail typically do not proceed to elaborate as to the particular shape the blackmail offense should take. The implication is that once the cost or externality that justifies blackmail’s prohibition has been identified, the definition of blackmail should be whatever minimizes (or produces the socially optimal level of) that cost. The perspective operates on the level of curtailing social harms rather than responding to individual acts of wrongdoing. Among theorists of this variety, Joseph Isenbergh stands out in offering an account of blackmail that specifies which particular cases of blackmail should be criminalized and which should not.60 He attempts to determine “what information is more valuable kept private and what information is more valuable disclosed.”61 Describing the threat recipient, blackmailer, and third party entitled to the information as A, B, and C, respectively, Isenbergh suggests the possibility of limiting the blackmail prohibition to two sets of cases: “1) information, however acquired, held by B concerning a prosecutable crime or tort committed by A against C; and 2) information acquired by B outside a prior course of dealing with A.”62 Isenbergh later elaborates on the second category, pointing out that the relevant distinction is “between information already held by B (or obtained fortuitously) and information generated by B’s special efforts for the purpose of blackmail.”63 Isenbergh then posits that situations where A and B “have a pre-existing relationship” are more likely to involve “information obtained fortuitously” in the course of the relationship, whereas situations where A and B have “no prior course of dealing” are more likely to involve “information deliberately farmed” and should hence be discouraged via legal sanction.64 Even for the cases Isenbergh recognizes as undesirable blackmail, however, it is worth noting that he does not advocate direct criminalization as the best legal response. Rather, Isenbergh would favor making the blackmail transaction legally unenforceable and, in the first category of cases, also making B legally complicit in A’s criminal or tortious wrongdoing.65 D. Theories of Blackmail as Noncrime: The Abolitionist Position A final possible response to the blackmail paradox is to conclude that its only proper resolution is to decriminalize blackmail, thereby eliminating the paradox. Some libertarian theorists have defended the position that

60. See Joseph Isenbergh, Blackmail From A to C, 141 U. PA. L. REV. 1905, 1907 (1993) (arguing that any justification for blackmail must “lie in the particular nature of information” and that for this reason it makes sense to criminalize certain forms of blackmail and not others). 61. Id. at 1927. 62. Id. at 1908. 63. Id. at 1929. 64. Id. at 1930. 65. Id. at 1928–29.

2010]


307

criminalizing threats to engage in lawful activity is an impermissible infringement on the freedom to engage in voluntary transactions, hence an unjustifiable exercise of criminal authority.66 Russell Christopher has offered a variation on this claim by introducing the concept of “meta-blackmail”: the conditional threat to make a blackmail threat.67 Christopher claims that there is no clear way to determine how meta-blackmail should be treated relative to blackmail itself—whether it should be punished more seriously, less seriously, or the same—and asserts that therefore the only way to avoid logical inconsistency, or at least even thornier puzzles than the paradox of blackmail itself, is to decriminalize blackmail (and hence meta-blackmail also).68 Mitchell Berman has argued against the soundness of Christopher’s logic.69 We need not concern ourselves here with the persuasiveness of Christopher’s account as an analytical matter because our project is to determine whether that account (or any other) accords with common moral sensibilities. In this case, the abolitionist position is easily testable: if subjects reject the prospect of punishment in all scenarios of putative blackmail, then their moral intuitions would track the conclusion that blackmail should be abolished, and if not, then lay intuitions would contradict the abolitionist proposal. It should also be noted, however, that a disagreement between lay intuitions and these accounts would not necessarily undercut the relevant theorists on their own terms. The libertarian position rests on a broader un-

66. See, e.g., 1 MURRAY N. ROTHBARD, MAN, ECONOMY, AND STATE 443 n.49 (1970) (“[B]lackmail would not be illegal in the free society. For blackmail is the receipt of money in exchange for the service of not publicizing certain information about the other person.”); Walter Block, Berman on Blackmail: Taking Motives Fervently, 3 FLA. ST. U. BUS. REV. 57, 61–62 (2003) (defining the libertarian view of blackmail as criminalizing something that the blackmailer has the right to do); Walter Block, The Case for De-criminalizing Blackmail: A Reply to Lindgren and Campbell, 24 W. ST. U. L. REV. 225, 225–26 (1997) (discussing how a transaction where one refrains from gossip for consideration from another party should be legal); Eric Mack, In Defense of Blackmail, 41 PHIL. STUD. 273, 273–74 (1982) (arguing that blackmail should not be prevented by the police power of the state); Ronald Joseph Scalise, Jr., Comment, Blackmail, Legality, and Liberalism, 74 TUL. L. REV. 1483, 1506 (2000) (“In a liberal legal system, all voluntary actions between consenting adults are allowable.”). 67. See Russell L. Christopher, Meta-Blackmail, 94 GEO. L.J. 739, 746 (2006) [hereinafter Christopher, Meta-Blackmail] (originating the concept of meta-blackmail); Russell L. Christopher, The Trilemma of Meta-Blackmail: Is Conditionally Threatening Blackmail Worse, the Same, or Better Than Blackmail Itself?, 94 GEO. L.J. 813, 813 (2006) (asking whether meta-blackmail may be more severe than simple blackmail). 68. See Christopher, Meta-Blackmail, supra note 67, at 747–48 (“Resolving the trilemma of meta-blackmail either forces the decriminalization of blackmail or adds considerably to the already difficult puzzles to be surmounted in justifying the criminalization of blackmail.”). 69. See Berman, supra note 6 (manuscript at 41–43) (arguing for the existence of a basis for differentiating meta-blackmail and blackmail); Berman, Meta-Blackmail, supra note 9, at 788 (arguing that the meta-blackmail “conceit” does not properly address the “widely and deeply held” opinion that some conditional threats to perform legal acts are properly criminalized).

308

Texas Law Review

[Vol. 89:291

derstanding that the proper scope of criminalization is very narrow and fails to justify a prohibition against blackmail even if blackmail is a moral wrong.70 Christopher does not purport to offer a moral refutation of the criminalization of blackmail, but what he says is a logical one.71 II.

Statutory Approaches

While each U.S. jurisdiction has a criminal provision prohibiting traditional blackmail,72 there is no single statutory approach used by a majority of states. Even within any given approach, a close comparison of any two blackmail statutes is likely to reveal some differences. However, the range of differences might be summarized as moving along two dimensions: first, the breadth of the range of demands criminalized;73 second, the breadth of the exceptions (or special defenses) to the crime. On the first dimension, blackmail statutes can be categorized as either having a broad range of prohibited demands74 or a narrow range.75 “Narrow

70. See Berman, supra note 6 (manuscript at 36–38) (noting the basis for the libertarian position and asserting that it “rests on a fairly straightforward, easily articulated and understood, major premise that the overwhelming majority of contemporary theorists of the criminal law simply reject”). 71. See Christopher, Meta-Blackmail, supra note 67, at 784–85 (arguing that “[c]riminalizing blackmail violates intuitions that are more compelling than the intuition that blackmail is properly criminalized”). 72. The standard blackmail case is one in which an actor threatens to disclose a damaging secret if the victim does not pay her some amount of money. Terminology varies from jurisdiction to jurisdiction; most prohibit the blackmail offense via a statute covering “criminal coercion,” “extortion,” “intimidation,” “threats,” or a similar term. 73. Interestingly, statutes also vary on the breadth of the range of prohibited threats, as opposed to the range of demands. All prohibit threats to disclose damaging secrets or expose a committed crime; many also criminalize threats to injure the victim or her property, to impugn the character of the victim or of some third party, to commit a crime, etc. However, these distinctions are irrelevant in the context of this study; it will suffice to note that all jurisdictions criminalize the threat inherent to traditional blackmail. 74. Thirty-three jurisdictions have broad ranges of prohibited demands: Alaska, Arkansas, Colorado, Connecticut, Florida, Hawaii, Illinois, Indiana, Iowa, Kansas, Kentucky, Louisiana, Maryland, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Montana, New Jersey, New Mexico, New York, North Carolina, North Dakota, Oklahoma, Pennsylvania, Rhode Island, Tennessee, Vermont, Washington, Wisconsin, Wyoming, and the District of Columbia. See ALASKA STAT. §§ 11.41.520, .530 (2008); ARK. CODE ANN. §§ 5-13-208, 5-36-101, -103 (2006); COLO. REV. STAT. ANN. § 18-3-207 (West 2009); CONN. GEN. STAT. ANN. §§ 53a-119, -192 (West 2007); FLA. STAT. ANN. § 836.05 (West 2005); HAW. REV. STAT. ANN. § 707-764 (LexisNexis 2007); 720 ILL. COMP. STAT. ANN. 5/12-6 (West 2007); IND. CODE § 35-45-2-1 (2004); IOWA CODE ANN. § 711.4 (West 2009); KAN. STAT. ANN. § 21-3428 (2007); KY. REV. STAT. ANN. §§ 509.080, 514.080 (LexisNexis 2008); LA. REV. STAT. ANN. § 14:66 (2007); MD. CODE ANN., CRIM. LAW §§ 3-701 to 3-708 (LexisNexis 2002); MASS. GEN. LAWS ANN. ch. 265, § 25 (West 2000); MICH. COMP. LAWS ANN. § 750.213 (West 2009); MINN. STAT. ANN. §§ 609.27–.275 (West 2009); MISS. CODE ANN. § 97-3-82 (West 2006); MO. ANN. STAT. § 570.010 (West 1999); MONT. CODE ANN. §§ 45-2-101, 45-5-305(1)(f), 45-6-301 (2008); N.J. STAT. ANN. §§ 2C:13-5, 2C:20-5 (West 2005); N.M. STAT. ANN. § 30-16-9 (LexisNexis 2000); N.Y. PENAL LAW §§ 135.60, 155.05 (McKinney 2008); N.C. GEN. STAT. §§ 14-118, 118.4 (2007); N.D. CENT. CODE §§ 12.1-17-06,

2010]


309

definition” statutes address only threats made to obtain property or pecuniary value. “Broad definition” blackmail statutes cover those threats and also threats made to coerce action on the part of the victim or some third party.76 The second dimension of blackmail is somewhat more complex: statutes can be categorized as having broad exceptions, narrow exceptions, or no exceptions. “Broad exception” statutes generally provide a form of goodfaith defense, holding that the blackmailer can escape liability where she is acting with the limited purpose of making the other party correct a wrong, desist from misbehavior, refrain from taking responsibility for which she is not qualified, or other similar situations.77 These statutes commonly impose

12.1-23-02, -23-10(12) (1997); OKLA. STAT. ANN. tit. 21, § 1488 (West 2003); 18 PA. CONS. STAT. ANN. §§ 2906, 3923 (West 2009); R.I. GEN. LAWS § 11-42-2 (2002); TENN. CODE ANN. § 39-14112 (2006); VT. STAT. ANN. tit. 13, § 1701 (2002); WASH. REV. CODE ANN. §§ 9A.04.110(27), 9A.56.110–.130 (West 2003); WIS. STAT. § 943.30–.31 (2009); WYO. STAT. ANN. § 6-2-402 (2009); D.C. CODE §§ 22-3201(4), -3252(a) (2001). 75. Nineteen jurisdictions have narrow ranges of prohibited demands: Alabama, Arizona, California, Delaware, Georgia, Idaho, Maine, Nebraska, Nevada, New Hampshire, Ohio, Oregon, South Carolina, South Dakota, Texas, Utah, Virginia, West Virginia, and the federal government. See ALA. CODE §§ 13A-8-1(13), 13A-8-13 to -15 (LexisNexis 2005); ARIZ. REV. STAT. ANN. § 131804 (2009); CAL. PENAL CODE §§ 518–19 (West 2008); DEL. CODE ANN. tit. 11, §§ 791, 846–47 (2007); GA. CODE ANN. § 16-8-16 (2007); IDAHO CODE ANN. § 18-2403 (2008); ME. REV. STAT. ANN. tit. 17-A, § 355 (2006); NEB. REV. STAT. § 28-513 (2008); NEV. REV. STAT. § 207.190 (2007); N.H. REV. STAT. ANN. § 637:5 (2007); OHIO REV. CODE ANN. § 2905.11–.12 (West 2006); OR. REV. STAT. §§ 163.275, 164.075 (2007); S.C. CODE ANN. § 16-17-640 (2005); S.D. CODIFIED LAWS § 22-30A-4 (2006); TEX. PENAL CODE ANN. §§ 31.01, .03 (West 2009) (note that Texas utilizes a common law duress offense to punish acts equivalent to blackmail); UTAH CODE ANN. § 76-6-406 (LexisNexis 2008); VA. CODE ANN. § 18.2-59 (2009); W. VA. CODE ANN. § 61-2-13 (LexisNexis 2005); 18 U.S.C. § 873 (2006). Note that some of these jurisdictions also have a very limited statute prohibiting the coercion of illegal action via threats. See, e.g., ALA. CODE § 13A-625; NEV. REV. STAT. § 207.190 (covering mostly classic extortion, i.e., threats of unlawful behavior); id. § 205.320 (covering threats to obtain property). While this may technically be approaching our definition of “broad ranges,” the illegal-action limitation makes the statute so narrow as to not be comparable with the broad-range statutes. 76. Most statutes recognize threats to harm or otherwise wrong a third party as blackmail. An example would be B telling V that he will harm J (V’s brother) if V does not pay. See, e.g., ALASKA STAT. § 11.41.520; ARIZ. REV. STAT. ANN. § 13-1804; CONN. GEN. STAT. ANN. § 53a-192; DEL. CODE ANN. tit. 11, § 846. 77. See, e.g., MODEL PENAL CODE § 212.5(1)(d) (1962); cf. N.D. CENT. CODE § 12.1-17-06 (defining criminal coercion, including affirmative defenses). Twenty jurisdictions have broad exceptions to the blackmail offense: Alaska, California, Connecticut, Florida, Hawaii, Indiana, Kentucky, Maryland, Massachusetts, Michigan, Mississippi, New Jersey, New Mexico, New York, North Carolina, North Dakota, Pennsylvania, Rhode Island, Vermont, and Washington. ALASKA STAT. §§ 11.41.520, .530; CAL. PENAL CODE §§ 518–19; CONN. GEN. STAT. ANN. §§ 53a-119, 53a192; HAW. REV. STAT. ANN. § 707-769; KY. REV. STAT. ANN. §§ 509.080, 514.080; N.J. STAT. ANN. §§ 2C:20-5, 2C:13-5; N.Y. PENAL LAW § 135.75; N.C. GEN. STAT. § 14-118; N.D. CENT. CODE §§ 12.1-17-06, 12.1-23-09; 18 PA. CONS. STAT. ANN. §§ 2906, 3923; WASH. REV. CODE ANN. §§ 9A.56.110–.130, 9A.04.110(27). This group includes jurisdictions that do not explicitly provide a good-faith defense but whose statutory language seemingly incorporates a bad-faith requirement into the offense definition itself. For example, the California extortion statute only criminalizes “the obtaining of property from

310

Texas Law Review

[Vol. 89:291

additional requirements on the actor and her behavior before she can make use of the exception.78 “Narrow exception” statutes generally hold that the blackmailer may be excepted only if she is acting to recover restitution for harm done or to recover compensation for property taken or services rendered.79 Finally, “no exception” statutes provide no explicit provisions recognizing exceptions to the blackmail offense.80 another, with his consent, or the obtaining of an official act of a public officer, induced by a wrongful use of force or fear, or under color of official right.” CAL. PENAL CODE § 518 (emphasis added). This would permit the obtaining of property from another by a nonwrongful use of force or fear, as where the actor’s motivation is to make another person right a previous wrong, stop creating harm, or disgorge stolen or otherwise unlawfully possessed property. Similar provisions exist in various “broad exception” statutes. See FLA. STAT. ANN. § 836.05; IND. CODE § 35-45-2-1; MD. CODE ANN., CRIM. LAW §§ 3-701 to -708; MASS. GEN. LAWS ANN. ch. 265, § 25; MICH. COMP. LAWS ANN. § 750.213; MISS. CODE. ANN. § 97-3-82; N.M. STAT. ANN. § 30-16-9; N.C. GEN. STAT. § 14-118; R.I. GEN. LAWS § 11-42-2; VT. STAT. ANN. tit. 13, § 1701. 78. For example, the Model Penal Code requires that an actor limit her purpose to compelling the “good” action, that the action coerced be reasonably related to the circumstances involved, and that the actor believe the accusation or secret revealed to be true. MODEL PENAL CODE § 212.5(1). 79. Twenty-two jurisdictions have narrow exceptions to the blackmail offense: Alabama, Arizona, Arkansas, Colorado, Delaware, Georgia, Idaho, Illinois, Iowa, Maine, Missouri, Montana, Nebraska, New Hampshire, South Carolina, South Dakota, Tennessee, Texas, Utah, Wisconsin, Wyoming, and the District of Columbia. ALA. CODE §§ 13A-6-25, 13A-8-1(13), 13A-8-13 to -15; ARIZ. REV. STAT. ANN. § 13-1804; ARK. CODE ANN. §§ 5-13-208, 5-36-101, 5-36-103; COLO. REV. STAT. ANN. § 18-3-207; DEL. CODE ANN. tit. 11, §§ 792, 847; GA. CODE ANN. § 16-8-16; IDAHO CODE ANN. § 18-2403; 720 ILL. COMP. STAT. ANN. 5/12-6; IOWA CODE ANN. § 711.4; ME. REV. STAT. ANN. tit. 17-A, §§ 355, 361; MO. ANN. STAT. § 570.010; MONT. CODE ANN. §§ 45-2101, 45-6-301, 45-5-305(1)(f); NEB. REV. STAT. § 28-513; N.H. REV. STAT. ANN. § 637:5; N.M. STAT. ANN. § 30-16-9; S.D. CODIFIED LAWS § 22-30A-4; TENN. CODE ANN. § 39-14-112; TEX. PENAL CODE ANN. §§ 31.01, 31.03; UTAH CODE ANN. § 76-6-406; WIS. STAT. §§ 943.30–.31; D.C. CODE §§ 22-3201(4), 22-3252(a). This group includes jurisdictions without explicit exception clauses but with statutory language seemingly designed to provide an exception from prosecution for cases in which the actor was attempting to recover property to which he had a legal entitlement. Examples include jurisdictions such as Colorado, where the statute only criminalizes blackmail committed “without legal authority.” COLO. REV. STAT. ANN. § 18-3-207(1)(a); see also 720 ILL. COMP. STAT. ANN. 5/12-6; MONT. CODE ANN. §§ 45-2-101, 45-6-301, 45-5-305(1)(f); WIS. STAT. §§ 943.30–.31. Presumably, one would have legal authority to recover taken property or recover compensation for past harm. Another example is the District of Columbia, where the offense definition criminalizes blackmailing with intent to obtain “property of another,” defined by statute as “any property in which a government or a person other than the accused has an interest which the accused is not privileged to interfere with or infringe upon without consent . . . .” D.C. CODE §§ 22-3252(a), 22-3201(4). An actor may be privileged to infringe on property owed to him; in such a case it would not be “property of another,” and the actor will not be liable. Similar provisions exist in many statutes categorized as having narrow exceptions. See IDAHO CODE ANN. § 18-2403; ME. REV. STAT. ANN. tit. 17-A, §§ 355, 361; N.H. REV. STAT. ANN. § 637:5; S.C. CODE ANN. § 16-17-640; S.D. CODIFIED LAWS § 22-30A-4; TEX. PENAL CODE ANN. §§ 31.01, 31.03; UTAH CODE ANN. § 76-6406; WYO. STAT. ANN. § 6-2-402. 80. Ten jurisdictions have no exceptions to the blackmail offense: Kansas, Louisiana, Minnesota, Nevada, Oklahoma, Ohio, Oregon, Virginia, West Virginia, and the federal government. KAN. STAT. ANN. § 21-3428; LA. REV. STAT. ANN. § 14:66; MINN. STAT. ANN. §§ 609.27–.275; NEV. REV. STAT. §§ 205.320, 207.190; OKLA. STAT. ANN. tit. 21, § 1488; OHIO REV. CODE ANN. § 2905.11–.12 (West 2006); OR. REV. STAT. §§ 163.275, 164.075 (2007); VA. CODE ANN. § 18.259; W. VA. CODE ANN. § 61-2-13; 18 U.S.C. § 873.

2010]


311

Relying upon these two dimensions of blackmail formulations creates four major categories into which the fifty-two American blackmail statutes fall:81 broad–broad, broad–narrow, narrow–narrow, and “other.” (There are narrow–broad statutes—a narrow definition of the crime, with broad exceptions—but the narrow definition means that the broad exceptions are never really used, so they are effectively the same in their operation as the narrow–narrow statutes.) Nineteen jurisdictions follow the Model Penal Code’s (MPC) broad– broad approach to blackmail by prohibiting threats made to coerce action or to take property and providing either explicit or implicit exceptions to the crime for actors who commit the offense in the course of an attempt to make the victim behave in a way reasonably related to the circumstances that were the subject of the threat.82 While these jurisdictions generally follow the MPC’s statutory language, there is some variation; Washington, for example, prohibits seeking “property or services” but specifically mentions sexual favors as being included in the definition of “services.”83 Other jurisdictions are not so explicit. Additionally, there is some variation in the defined exceptions to the crime. Most broad–broad jurisdictions employ the MPC’s formulation, but some limit the applicability of the exception to certain situations, and others (most significantly North Dakota) dramatically broaden the MPC’s exception. Ten jurisdictions take the broad–narrow approach, criminalizing threats designed to coerce action or to take property but providing an exception only

81. The fifty-two statutes are the codes of each of the fifty states plus the federal code and the District of Columbia code. 82. Alaska, Connecticut, Florida, Hawaii, Indiana, Kentucky, Maryland, Massachusetts, Michigan, Mississippi, New Jersey, New Mexico, New York, North Carolina, North Dakota, Pennsylvania, Rhode Island, Vermont, and Washington. See supra notes 74, 77 and accompanying text. A typical broad–broad statute would be similar to Pennsylvania’s: (a) Offense defined.—A person is guilty of criminal coercion, if, with intent unlawfully to restrict freedom of action of another to the detriment of the other, he threatens to: (1) commit any criminal offense; (2) accuse anyone of a criminal offense; (3) expose any secret tending to subject any person to hatred, contempt or ridicule; or (4) take or withhold action as an official, or cause an official to take or withhold action. (b) Defense.—It is a defense to prosecution based on paragraphs (a)(2), (a)(3) or (a)(4) of this section that the actor believed the accusation or secret to be true or the proposed official action justified and that his intent was limited to compelling the other to behave in a way reasonably related to the circumstances which were the subject of the accusation, exposure or proposed official action, as by desisting from further misbehavior, making good a wrong done, refraining from taking any action or responsibility for which the actor believes the other disqualified. 18 PA. CONS. STAT. ANN. § 2906. 83. WASH. REV. CODE ANN. § 9A.56.110.

312

Texas Law Review

[Vol. 89:291

to actors who make the threat in order to recover restitution for harm done or to gain compensation for services rendered or property owed.84 These statutes exhibit significantly more terminological and structural variation than do the broad–broad statutes; some are simply MPC-style provisions with narrower exceptions, some are uniquely drafted but part of a modern code structure, and others are common law-style provisions.85 Some broad– narrow statutes ban an extensive list of threats,86 while others are much more limited.87 Nonetheless, these statutes are appropriately grouped because all prohibit threats seeking action or property but only provide an offense exception if the actor is seeking property to which she has some legal right. Thirteen jurisdictions take the narrow–narrow approach, criminalizing threats made to gain property and providing an offense exception only where the actor makes an otherwise-prohibited threat in order to recover restitution for harm done or to gain compensation for services rendered or property owed.88 As with the broad–narrow jurisdictions, these statutes exhibit 84. Arkansas, Colorado, Illinois, Iowa, Missouri, Montana, Tennessee, Wisconsin, Wyoming, and the District of Columbia. See supra notes 74, 79 and accompanying text. Broad–narrow statutes can be constructed in a number of ways. One of the simplest is Tennessee’s: (a) A person commits extortion who uses coercion upon another person with the intent to: (1) Obtain property, services, any advantage or immunity; or (2) Restrict unlawfully another's freedom of action. (b) It is an affirmative defense to prosecution for extortion that the person reasonably claimed: (1) Appropriate restitution or appropriate indemnification for harm done; or (2) Appropriate compensation for property or lawful services. TENN. CODE ANN. § 39-14-112; see also D.C. CODE §§ 22-3252(a), 22-3201(4); MICH. COMP. LAWS ANN. § 750.213 (requiring the threat to be “malicious” to constitute a violation). 85. See, e.g., ARIZ. REV. STAT. ANN. § 13-1804 (MPC style); MINN. STAT. ANN. §§ 609.27– 609.275 (modern structure code); MASS. GEN. LAWS ANN. ch. 265, § 25 (common law style). 86. See, e.g., 720 ILL. COMP. STAT. ANN. 5/12-6(a); IOWA CODE ANN. § 711.4. 87. See, e.g., MICH. COMP. LAWS ANN. § 750.213; TENN. CODE ANN. § 39-14-112. 88. Alabama, Arizona, Delaware, Georgia, Idaho, Maine, Nebraska, New Hampshire, South Carolina, South Dakota, Texas, and Utah are “true” narrow–narrow states, while California has a narrow prohibition but seemingly broad exceptions. See supra notes 75, 77, 79 and accompanying text. In practice, however, the distinction between narrow–narrow and narrow–broad statutes appears to be irrelevant; if the offense only makes seeking property via blackmail a crime, an exception that goes beyond rightful property recovery (the essence of “narrow exceptions”) will never have any effect. Narrow–narrow statutes, like broad–narrow statutes, do not share a general pattern as do most broad–broad statutes. However, Arizona’s theft by extortion statute is typical of those jurisdictions with a narrow demand language and a narrow affirmative defense: A. A person commits theft by extortion by knowingly obtaining or seeking to obtain property or services by means of a threat to do in the future any of the following: 1. Cause physical injury to anyone by means of a deadly weapon or dangerous instrument. 2. Cause physical injury to anyone except as provided in paragraph 1 of this subsection. 3. Cause damage to property. 4. Engage in other conduct constituting an offense. 5. Accuse anyone of a crime or bring criminal charges against anyone.

2010]


313

considerable variation in statutory language and organization. Many are codified as extortion statutes, but they are intended to cover the traditional blackmail crime also.89 In the “other” category are ten jurisdictions that appear to have no exceptions to their blackmail laws.90 Four have broad prohibitions,91 and six have narrow prohibitions.92 III. Testing Community Views To test which of the theories and statutory schemes best capture lay intuitions about the conduct that should be criminalized, subjects were given a series of scenarios designed to focus on the differences among the theories. As to each scenario, a test document asked whether such conduct should be criminalized. To be sure that subjects were perceiving the scenarios as intended, a second test document performed a “manipulation check,” asking for details about subjects’ perception of each scenario—specifically questions testing what the subject perceived with regard to each of the factors that 6. Expose a secret or an asserted fact, whether true or false, tending to subject anyone to hatred, contempt or ridicule or to impair the person’s credit or business. 7. Take or withhold action as a public servant or cause a public servant to take or withhold action. 8. Cause anyone to part with any property. B. It is an affirmative defense to a prosecution under subsection A, paragraph 5, 6 or 7 that the property obtained by threat of the accusation, exposure, lawsuit or other invocation of official action was lawfully claimed either as: 1. Restitution or indemnification for harm done under circumstances to which the accusation, exposure, lawsuit or other official action relates. 2. Compensation for property that was lawfully obtained or for lawful services. ARIZ. REV. STAT. ANN. § 13-1804. Other statutes achieve the same ends via different means. See, e.g., GA. CODE ANN. § 16-8-16; LA. REV. STAT. ANN. § 14:66. 89. See, e.g., ALA. CODE § 13A-8-15 (“Extortion by means of a threat . . . constitutes extortion in the second degree.”). 90. Jurisdictions with “no exception” blackmail statutes have varied approaches. Kansas’s blackmail statute is simple: “Blackmail is gaining or attempting to gain anything of value or compelling another to act against such person’s will, by threatening to communicate accusations or statements about any person that would subject such person or any other person to public ridicule, contempt or degradation.” KAN. STAT. ANN. § 21-3428 (2007). Other “no exception” statutes vary. See, e.g., OKLA. STAT. ANN. tit. 21, § 60-1488 (West 2003) (codifying two exclusive components to Oklahoma’s blackmail law—accusing or threatening to accuse a person of a crime or exposing or threatening to expose facts that would “subject such person to the ridicule or contempt of society”— but not recognizing any exceptions to the statute); 18 U.S.C. § 873 (2006) (declaring no statutory exceptions to federal blackmail law if an individual receives some value for the threat of informing or as consideration for not informing of any violation of the law of the United States). 91. KAN. STAT. ANN. § 21-3428; LA. REV. STAT. ANN. § 14:66; MINN. STAT. ANN. §§ 609.27– 609.275; OKLA. STAT. ANN. tit. 21, § 60-1488. 92. NEV. REV. STAT. § 205.320 (2007); OHIO REV. CODE ANN. § 2905.11–.12 (West 2006); OR. REV. STAT. ANN. § 164.075 (2007); 18 U.S.C. § 873 (2006); VA. CODE ANN. § 18.2-59 (2009); W. VA. CODE ANN. § 61-2-13 (LexisNexis 2005).

314

Texas Law Review

[Vol. 89:291

were being manipulated to test the different theories. The subjects’ criminalization responses were then compared to those predicted by each theory and each statutory approach to determine which best reflected the subjects’ views. A. Design and Methodology The subjects were 129 men and women,93 recruited through flyers and an e-mail listserv, who were brought into a quiet laboratory and completed the study in return for $4. Subjects were run in small groups (usually one to three per group). Each subject was tested individually, completing the study at his or her own pace. Subjects ranged in age from eighteen to fifty-eight and as a group were ethnically, educationally, and economically diverse.94 A series of pilot tests revealed that the order in which the two test documents were given had no effect on results, so all subjects were given the manipulation-check questionnaire first, then the criminalization questionnaire. Each questionnaire presented subjects with the same eleven scenarios (reproduced in Table 1) that were constructed to highlight the differences among the theories being tested. We administered two versions of the questionnaire, which varied the order of presentation of the eleven scenarios. Presentation order did not significantly affect any of the results we report. As is apparent, each of the scenarios involves two main characters, Victor and Brian. In each case, Brian is the potential blackmailer who threatens Victor, the potential victim, and demands, in return for not carrying out the threat, some action or other compensation. The threat and the demand in each scenario are different, and each scenario generates a different pattern with respect to whether it satisfies the criteria of each of the theories. Table 1. Text of Scenarios 1. Pay or Report Crime. Brian learns that Victor has killed a man and tells Victor he will report the crime to the authorities unless Victor pays him $1,000. (continued)

93. Seventy-seven women, fifty-one men, and one subject unspecified. 94. Average age was 25.0, with a standard deviation of 7.9. Ethnicity of the subjects consisted of 58 white, 41 African-American, 15 Asian, 8 Latino, 4 multi-ethnic, and 3 Native American. Educationally, the subjects consisted of 7 high school, 69 some college, 46 college degree, 5 masters degree, and 2 professional degree, and their self-reported household incomes ranged from $10,000 to $175,000 (75th percentile = $65,000; median = $20,000; 25th percentile = $10,000).

2010]


315

Table 1 (cont.). Text of Scenarios 2. Pay or Face Lawsuit. Victor, driving negligently, seriously damages Brian’s car in an accident. Brian tells Victor that he will sue him in court, where he can collect the cost of repairs, unless Victor pays him the $1,000 that the repairs will cost. 3. Sober Work or Expose Drinking. Brian and Victor both work in a factory. Brian discovers that Victor has a drinking problem, which not only explains his frequent absences and tardiness but might also create a risk in the workplace. Victor is worried that if management finds out, he will lose his job under the company’s “zero tolerance” policy, which mandates dismissal of anyone found to have an existing drinking problem. Brian says he will report Victor’s drinking problem to management unless Victor promises to show up sober every day to work, join an alcohol abuse treatment program to avoid recurrence, and make a donation to a charity fighting alcohol abuse. 4. Pay or Reveal Recipe. Victor owns a bakery known for its cupcakes, which are very popular despite their high price. Brian discovers that the cupcakes are actually made using a cheap store-bought cake mix and frosting. Though Victor has never made any false claims about how his cupcakes are made, he knows his business will be ruined if the truth is revealed. Brian threatens to make the cupcake “recipe” public unless Victor pays him $10,000. 5. Pay or Publish Book. Brian, a literature professor, has spent considerable time conducting research for a biography of Victor, a famous author. His research has turned up information that would destroy Victor’s reputation. Despite what he has learned, Brian admires Victor and does not wish to cause him harm, but he also thinks it would be inappropriate to publish a biography that does not accurately present what he knows. Accordingly, he contacts Victor and says he will discontinue his biography project if Victor pays him the $4,000 that will properly compensate Brian for his expenses and the work he has already done. (continued)

316

Texas Law Review

[Vol. 89:291

Table 1 (cont.). Text of Scenarios 6. Pay or Expose Cheating. Brian and Victor are taking the examination to enter the police academy. All applicants sign a promise not to cheat and to report those who do. Authorities aggressively prosecute cheating but only if there is hard evidence, such as the “cheat sheet” used during the exam. During the exam, Brian sees and picks up Victor’s “cheat sheet.” He says he will give it to the authorities unless Victor pays him $500. If Victor pays, Brian will give the “cheat sheet” back, and no action of any kind against Victor will be possible. 7. Withdraw or Expose Cheating. Brian and Victor are taking the examination to enter the police academy. All applicants sign a promise not to cheat and to report those who do. Authorities aggressively prosecute cheating but only if there is hard evidence, such as the “cheat sheet” used during the exam. During the exam, Brian sees and picks up Victor’s “cheat sheet.” Victor is remorseful about the cheating, but does not want to face legal action. Brian says he will expose the cheating unless Victor withdraws his application to the police force, so that he cannot actually benefit from this instance of cheating. If Victor does so, Brian will give back the cheat sheet, and no action of any kind against Victor will be possible. 8. Cut Tree or Publish Photos. Brian, Victor’s neighbor, thinks that Victor’s expensive and exotic tree is an eyesore. Brian has asked Victor to cut the tree down but Victor has refused to do so. Using a special telephoto lens, Brian takes photos of Victor in his home having sexual intercourse and tells Victor that he will post them on the Internet unless Victor agrees to cut down the offending tree. If Victor does cut it down, Brian will hand over the negatives and the only copy of the photos. (continued)

2010]


317

Table 1 (cont.). Text of Scenarios 9. Pay or Report Smoking. Brian and Victor work for a nonprofit antismoking organization. Teresa, the current head of the organization, has made clear that she expects all her employees to not smoke, even though this was not an explicit condition for their employment, because she views smokers as insufficiently committed to her organization’s mission. However, Teresa is permanently leaving the organization. Her replacement, Tina, does not care whether her employees smoke when they are outside the workplace. A few days before Teresa’s departure, Brian runs into Victor at a restaurant. Victor is halfway through a cigar. He asks Brian not to tell Teresa, because he knows that if she finds out, her last act as head of the organization will be to fire him. Victor tells Brian that he loves his job, though he also enjoys smoking on occasion. Brian says he will tell Teresa immediately unless Victor pays him $1,000. 10. Pay or Report Crime. Brian doesn’t like Victor. When he learns that Victor has killed a man, he tells Victor that he will report the crime to the authorities unless Victor pays him $1, as a way of forcing an admission of guilt that he might choose to use against Victor in the future. 11. Pay Penalty or Foreclose. Brian is the banker who oversees Victor’s home mortgage loan. He is also in love with Victor’s wife. Victor has missed several mortgage payments. Brian is authorized by his bosses to either arrange a refinancing of Victor’s loan, or to impose a substantial one-time penalty on Victor for missing past payments. Brian tells Victor that unless Victor pays the substantial penalty — which he assumes Victor cannot do — the bank will foreclose on Victor’s house. Brian believes if Victor loses his house, Victor’s wife will leave him.

The first and second scenarios are designed for quality-control purposes. The first scenario is a classic case of blackmail for which all theorists95 and all statutory schemes96 would impose blackmail liability. Scenario 2 provides an example of the reverse case, one in which all theorists and all statutory schemes would agree that no blackmail liability should exist.97 If a subject were to give an incorrect answer to either or both of these screening questions, that subject would be segregated from the bulk of the subjects and not included in the analysis of survey results. We are committed to putting

95. See supra subparts I(A)–(C). But see supra subpart I(D). 96. See supra notes 73–79 and accompanying text. 97. See supra subpart I(C).

318

Texas Law Review

[Vol. 89:291

the various theories and code-drafting approaches to a fair test, which should not include using subjects who cannot provide proper results for the clear cases presented in the two screening scenarios.98 To be sure that the subjects were in fact perceiving the scenarios in the way that each was intended, the manipulation-check questionnaire asked each subject whether he or she perceived certain facts or conclusions about each scenario, specifically those facts or conclusions that served as the criteria for each theory. The standard templates for each manipulation-check question for each of the four theories are set out in Table 2. (Recall that the Isenbergh economic theory made no claim that it was based in any part upon lay intuitions of justice, so there is no manipulation check for it as there is for each of the other theories, although we will later compare its liability preferences to the liability preferences of lay persons.)

98. As a result of this screening mechanism, thirty-five subjects were excluded from the analysis: twenty-one who varied from the predicted response to Scenario 1, twenty-nine who varied from the predicted response to Scenario 2, and fifteen who varied on both. For the fifteen who confounded both predictions, it is hard to see how their responses could indicate anything other than confusion, random answering, or malicious mischief, as any principled disagreement to the accepted result in the two cases would arise from different, and indeed opposing, views (abolitionists versus expansionists). In fact, as a group, those fifteen subjects’ overall responses were “indifferent” (i.e., not statistically significant relative to a neutral answer) for seven of the remaining nine responses, suggesting randomness. (The other two scenarios were Scenario 5 (Pay or Publish Book), for which the excluded subjects favored liability but the included subjects gave an indeterminate response, and Scenario 10 (Pay or Report Crime), where the excluded subjects favored no liability and the included subjects favored liability.) Those who “erred” on Scenario 1, rejecting liability where liability was predicted, might have been demonstrating an “abolitionist” position toward blackmail, thinking it should never be punished. Yet these respondents as a group also gave indifferent responses to seven of the other ten scenarios (including scenarios where respondents as a whole consistently rejected liability) and gave pro-liability responses to another two scenarios (Scenarios 4 (Pay or Reveal Recipe) and 8 (Cut Tree or Publish Photos)). In fact, the only other scenario for which this group decisively rejected liability was Scenario 10—the other “Pay or Report Crime” scenario. Again, this pattern of responses suggests arbitrariness or outright deception. Those who erred on Scenario 2 also gave indifferent responses to seven of the other ten scenarios (though not the same seven as for those who erred on Scenario 1). For the other three— Scenarios 5 (Pay or Publish Book), 6 (Pay or Expose Cheating), and 8 (Cut Tree or Publish Photos)—this group favored liability. The excluded subjects also fared very poorly on the manipulation checks, providing further reason to ignore their responses. Of the forty-four manipulation checks, those who “erred” on Scenario 1 gave indifferent responses to nineteen, and “wrong” (i.e., the opposite of predicted) answers to another six; those who erred on Scenario 2 also gave nineteen indifferent answers and six wrong ones, though they were not for the same sets of manipulation checks as the other group.

2010]


319

Table 2. Manipulation-Check Questions for Criteria for Each Theory a. Wrongful Intention. If Victor refuses Brian’s offer, and Brian carries out the threat, would his primary reason for acting be wrongful? (+ = liability) [a1: Berman] [& finding demand is substantial = a2: Katz] b. Offer to Violate Legal Duty. If another person knew what Brian knows, should the law require that the person do [or: forbid the person from doing] what Brian threatens to do to Victor, or face legal liability? (+ = liability) [Feinberg] c. Continuing Domination. If Victor agrees to the demand, would Brian retain the power to make additional demands based on the same threat on a future occasion? (+ & finding demand is substantial = liability) [Fletcher] d. Leveraging Another’s Influence. If an interested third party learned what Brian knows, would the third party’s reaction be to want something different from what Brian demands that Victor do? (+ = liability) [Lindgren] * Scenario #2 required a special d question because the standard d question above assumes that there might be an interested third party, but scenario 2 was constructed to not have one. Thus, we needed to confirm that the subjects perceived this to be the case and asked: Is there an interested third party who should more properly be exercising the authority that Brian is exercising when he threatens Victor? (If you think that Brian is threatening to use authority that properly belongs to him alone, select –3, –2, or –1.) (– = no liability)

However, each manipulation-check question was in fact individualized to reflect the facts of each scenario. Thus, for example, the manipulation check for the Wrongful Intention Theory for Scenario 1 was as follows. a. If Victor does not pay Brian the $1,000 and Brian reports Victor’s crime to the authorities, would Brian’s primary reason for reporting Victor’s crime be wrongful? –3 Clearly no

–2 Probably no

–1 Possibly no

+1 Possibly yes

+2 Probably yes

+3 Clearly yes

320

Texas Law Review

[Vol. 89:291

The same six-point, no–yes scale was used for all forty-four of these questions (one question for each of the four theories’ criteria for each of the eleven scenarios). Following these four manipulation-check questions, three additional questions were asked with regard to each scenario. One asked whether the threat was substantial—an additional element of two of the theories (Katz’s and Fletcher’s). The two other questions asked about the extent of the harm that would be caused if Brian did as he threatened and about the wrongfulness of the victim’s conduct that Brian was threatening to expose. Again, each of these three questions was customized to the facts of the scenario. Thus, the questions for Scenario 1 were as follows, with the response scales shown below. e. How substantial is a demand that another person pay $1,000? 0 No demand

1 Trivial demand

2 Somewhat substantial

3 Substantial

4 Very substantial

5 Extremely substantial

f. What is the extent of the harm that would be caused to Victor if Brian reported Victor’s crime to the authorities? 0 Not damaging

1 Only trivially damaging

2 Somewhat damaging

3 Damaging

4 Seriously damaging

5 Extremely damaging

g. How wrongfully did Victor behave by killing a man? 0 Not wrongful at all

1 Trivial

2 Somewhat wrongful

3 Wrongful

4 Seriously wrongful

5 Extremely wrongful

The last two questions were asked so we could test whether these variables might play a role in the subjects’ criminalization decisions. Notice that all three of these questions asked not merely for a binary response (e.g., wrongful or not wrongful) but for a quantitative measure of the factor, which we could then use to see if it correlated with subjects’ criminalization decisions.

2010]


321

The “correct answers”—those sought by the researchers to assure subjects were perceiving each scenario as intended—are set out in Table 3 below. Table 3. Manipulation Check “Correct Answers” #

Scenario

WI

BD

CD

UA

SD

1

Pay or Report Crime

+

+

+

+

≥2

2

Pay or Face Lawsuit

–

–

–

–*

≥2

3

Sober Work or Expose Drinking

–

–

–

+

≥2

4

Pay or Reveal Recipe

+

–

+

+

≥2

5

Pay or Publish Book

–

–

+

+

≥2

6

Pay or Expose Cheating

+

+

–

+

≥2

7

Withdraw or Expose Cheating

–

+

–

+

≥2

8

Cut Tree or Publish Photos

+

+

–

+

≥2

9

Pay or Report Smoking

+

–

–

+

≥2

10

Pay or Report Crime

+**

+

+**

+